One of the most common questions I encounter is the future of working from home in the aftermath of the COVID-19 pandemic. Audiences of all types are curious about the future magnitude, its impact on retail office space, traffic congestion, the health of downtown economies, and myriad other aspects of economic life. Has American work life been altered forever?

My instinct has always been to downplay the permanent shifts. People don’t simply change overnight and cities are enormous engines of productivity and economic activity. If tastes and incentives are the same post-pandemic, the outcomes will be the same as well. My view was also informed by the fact that during the pandemic the Bureau of Labor Statistics (BLS) added supplemental questions regarding working from home to the monthly Household Survey.

In September of 2020, it reported: “In August, 24.3 percent of employed persons teleworked because of the coronavirus pandemic, down from 26.4 percent in July. These data refer to employed persons who teleworked or worked at home for pay at some point in the last 4 weeks specifically because of the coronavirus pandemic.” If roughly 25 percent of employees worked from home at the peak of the pandemic, then 75 percent continued to report to the job site. If the vast majority of workers were unaffected, it struck me as unlikely that a minority – probably well below 20 percent in the end – would overturn the spatial structure of the economy.

So, imagine my surprise this past week when the BLS released the 2022 results for the American Time Use Survey (ATUS) and noted: “On days they worked, 34 percent of employed persons did some or all of their work at home and 69 percent of employed persons did some or all of their work at their workplace. On average, those who worked at home did so for 5.4 hours on days they worked, and those who worked at their workplace did so for 7.9 hours.”

Thirty-four percent! I did not want to believe that I was so far wrong. But those who were at home worked a lot less than those at the job site. I did believe that.

Of course, there are differences in the samples used in the ATUS and the monthly Household Survey and the precise way that questions are asked always matters. So it dawned on me that I might want to check into that. To do so, I took a look at the 2019 – i.e., pre-pandemic – ATUS. In 2019, the result was that 23.7 percent worked from home and did so for 3.3 hours. In contrast 81.9 percent worked at the workplace for 7.9 hours.

Inside the 2022 top-line numbers one finds patterns that are unsurprising. Work at home is more common for women (41 percent) than men (28 percent). It also rises steadily with educational attainment (for those aged 25 and older) from 8.0 percent for those without a high school diploma to 53.7 percent for those with a bachelor’s degree or greater. Average hours working at home also rise with education from 4.2 hours to 5.5 hours.

It seems to me that the real story is that without much notice the workplace culture was shifting well before the impact of COVID-19. Yes, since 2019, the fraction working at home is up by about 10 percentage points and an average of 2.1 hours, but it is a far less seismic shift than portrayed by simply looking at the 2022 ATUS.